Data Science newsletter – December 18, 2019

GROUP CURATION: N/A

BMW has been implementing next-level production processes throughout its plants around the world for quite some time now. From robots carrying parts around the compound, to exoskeletons for the workers, a number of innovative features have been added to the mix. Today, the Bavarian manufacturer announced the publishing of its AI algorithms used in production to be used by anyone interested. They are now available on the popular GitHub platform.

The algorithms published are mainly part of various AI applications, in particular in automated image recognition and image tagging. Making these publicly available allows software developers all over the world to view, change, use and improve the source code. “With the algorithms we are now publishing, the BMW Group has significantly reduced the development time for neural networks for autonomous transport systems and robots,” says Dirk Dreher, Head of Logistics Planning.

For a decade, digital health has been the supposed savior of the health care system, driving health care into a data-first, low-cost industry worthy of the 21st-century.

For that reason, investors have poured over $30 billion into digital health since 2011.

But almost a decade in, what material change can we point to in health care costs or the experience of the average patient? Are there companies that qualify as major disruptors? To me, the answer is no. And I call this “the Digital Health Conundrum.”

Oilfield service giant Schlumberger has inked a deal to expand the use of artificial intelligence technology in the oil patch.

In a statement, Schlumberger announced it had entered into an agreement the New York software company Dataiku.

Under the agreement, the two companies will work together to develop artificial intelligence products and services for Schlumberger’s exploration and production customers.

The techlash hasn’t died down – it’s just become normalized. Barely a day passes without a new scandal emerging, from questionable surveillance to racist AI algorithms. But it hasn’t all been bad: while negatives get a lot of attention (and so they should – the consequences of tech can be lethal, both societally and literally), there was still plenty to get excited about. And for those working in the data profession – as analysts, scientists, and engineers, there were several important trends that really helped to define where we are now from a purely practical perspective – as well as hinting at where we might go in the future.

With just a few weeks left to go of the year (and the decade!), let’s look at some of the key things that defined this year in the field of data science and data engineering.

Without a response, Rudy Espericuta, who was with Guebara and her children at the time, dialed 911. About 15 minutes later, after the fight had ended, a woman was rolled out on a stretcher and into an ambulance, her head bleeding from a cut suffered during the altercation.

Amid the scene, the robot continued to glide along its pre-programmed route, humming an intergalactic tune that could have been ripped from any low-budget sci-fi film. The almost 400-pound robot followed the park’s winding concrete from the basketball courts to the children’s splash zone, pausing every so often to tell visitors to “please keep the park clean.”

Rural communities face growing infrastructure problems like decaying water systems. And they have more limited access to amenities ranging from grocery stores to movie theaters, lower quality schools, and less access to high-speed internet.

Yet perhaps most daunting are the tremendous health disparities rural Americans face, in terms of both their own health and accessing care.

As a number of my recent studies indicate, these disparities may be exacerbated by insurance carriers and the networks they put together for their consumers.

How has the space gone from bad to worse while rates of multi-drug resistant infection and deaths caused by superbugs steadily increase? Well, it’s actually not entirely clear how many lives are claimed by lack of effective antibiotics. Patients who contract, or succumb to, a resistant infection are severely undercounted. Under the Affordable Care Act, hospitals must pay a penalty for each hospital-acquired-infection (HAI) occurring within their in-patient population. As a result, if a patient dies from a superbug contracted during a procedure such as surgery, the official cause of death may be instead listed as “Complications from Surgery.” Consistent and systemic undercounting of illnesses and deaths from resistant infections further discourages the development of new antibiotics as the number of patients who need these medicines may appear to be very small.

The failure of the market for new antibiotics has also been caused by several economic and commercial factors. Approval incentives were not the only policy included in the GAIN Act. There were also measures designed to promote stewardship, or appropriate use, of new antibiotics. In short, when a new antibiotic becomes available, it should only be used as a last resort to prevent new resistance from arising. This kind of responsible use is a good thing! But stewardship severely limits the number of patients who will receive a new antibiotic and, correspondingly, the potential sales volume.

In less than two decades, the Arctic Ocean is expected to be nearly ice-free in the late summer. That loss has great consequences. Arctic ice acts as a global refrigerator, reflecting back radiation from the sun. Moreover, water from the melting of ice stored on land in Greenland and other Arctic glaciers runs into the ocean, raising the average global sea level and causing coastal flooding. The Arctic Report Card 2019, produced Dec. 10 by the US National Oceanic and Atmospheric Administration, indicates that the pace of ice melt actually already matches the worst-case scenario from the recent IPCC report released just months ago.

As a marine ecologist, I never expected I would one day advocate that science should operate more like the tech industry.

This is not about “moving fast and breaking things.” For me, it is about openness.

Open software, both a driver and a result of Silicon Valley’s success, has been game-changing for me as a scientist. Its transformative power has improved my ability to analyze data and collaborate with other scientists.

Over the last few years, so-called “fake news” — purposefully untrue misinformation spread online — has become more and more of a concern. From extensive media coverage of the issue to government committees being set up for its investigation, fake news is at the top of the agenda — and more often than we’d like, on top of our newsfeeds.

But how does exposure to misinformation impact the way we respond to it? A new study, published in Psychological Science, suggests that the more we see it, the more we’re likely to spread it. And considering the fact that fake news is more likely to go viral than real news, this could have worrying implications.

Research has found that previously encountered information feels more “fluent” — in other words, we find it easier to process. This, in turn, gives it a “ring of truthfulness”, write Daniel Effron from London Business School and Medha Raj from the University of Southern California: repeated information feels true, even as we simultaneously acknowledge it’s not. And, the pair predicted, because our intuitions often drive our moral judgements, we may feel it less unethical to share frequently encountered misinformation, even if we know it’s false, simply because it has this “feeling” of truth.

A new PhD program at Washington University in St. Louis is looking at developing additional tools for use in social services and other areas dealing with human and social behavior: computer programs that can use this wealth of data to help humans better understand problems in the social sciences.

This is the inaugural semester of the Division of Computational and Data Sciences (DCDS), one of a few of its kind in the country, which focuses on turning the computational lens on social sciences.

In the past decade, the University of Notre Dame has made significant investments to create a strong foundation in new data-driven academic programs, including programs in applied and computational mathematics and statistics, business analytics, an online master’s degree in data science, a University-wide undergraduate minor in data science, and a proposed liberal arts computer science major, to name a few.

These programs complement and enhance existing undergraduate and graduate degree programs in computer science and computer engineering as well as a number of college and university centers and institutes devoted to the collection, curation and analysis of data to support decision-making in key problem domains across society.

The University will now take the next steps in these efforts, thanks to a transformative $25 million gift from alumnus Robert Lumpkins and his wife, Sara, a Saint Mary’s College graduate, to establish the Lucy Family Institute for Data & Society, an interdisciplinary, university-wide hub that will connect faculty, students and research across existing data science and analytics programs on campus while serving as a catalyst for future programmatic needs.

[Baskar] Ganapathysubramanian said the project’s immediate goal is to develop machine-learning theory and software tools that will allow rapid identification of organic thin film structures that enhance solar cell performance and are easy to manufacture. The broader goal is to demonstrate that machine learning can help rapidly design all kinds of technologies.

“We’re looking at a non-traditional way of doing machine learning – we’re doing science with machine learning,” said Sarkar, who studies machine learning and artificial intelligence. “Machine learning has been used to make your next Netflix recommendation. The new frontier is trying to see if machine learning can help engineers or scientists do engineering or science better.”

Tufts University has been awarded a three-year, $500,000 grant from the National Science Foundation to establish a TRIPODS Institute, which will broaden the university’s efforts in the development of new methods for securely extracting information from complex data sets and applying these techniques across diverse disciplines.

Tufts joins more than twenty other TRIPODS Institute universities (TRIPODS stands for Transdisciplinary Research In Principles of Data Science), which are a central piece of NSF’s Harnessing the Data Revolution initiative aimed at unlocking the potential of big data to drive innovation and discovery.

The T-TRIPODS Institute, as it will be called at Tufts, will bring together a cross-disciplinary team of faculty in mathematics, computer science, and electrical and computer engineering who will develop and share their expertise in data science with faculty and students in fields outside their own, including the classics, urban planning, and veterinary medicine.

“The Zwicky Transient Facility (ZTF) and IPAC at the California Institute of Technology announce the second ZTF Public Data Release. ZTF is an optical time-domain survey covering the northern sky visible from Palomar Observatory. This release builds upon the first data release to include products from (i) an additional 6 months of survey operations from the public portion of the survey, giving a total observation span of March 2018-June 2019, and (ii) data acquired under private survey time during the first ~3.4 months of the survey, spanning March 2018-June 2018.”

“FastMRI, a joint research collaboration between Facebook AI and NYU Langone Health to use AI to speed up magnetic resonance imaging (MRI) scans, is announcing a new open source data set from NYU Langone Health, along with baseline models and a newly expanded research paper to help the AI research community accelerate and broaden research in this area.”

A University of California, Irvine-led team of glaciologists has unveiled the most accurate portrait yet of the contours of the land beneath Antarctica’s ice sheet – and, by doing so, has helped identify which regions of the continent are going to be more, or less, vulnerable to future climate warming.

Highly anticipated by the global cryosphere and environmental science communities, the newly released Antarctica topography map, BedMachine, and related findings were published today in the journal Nature Geoscience.

Clustering and classification are two common machine learning methods for recognizing patterns in data. Here’s how they work and how they’re different.